Tracking URL not working for Accented Characters

[harrymsn0]

Folks,

URL Tracking breaks when there are Accented Characters.

Example of URL that does not work:
https://domain.com/auth/signin?utm_source=canal_proprietario&utm_medium=email&utm_campaign=1_botão2

I tried to fix it by editing app/lib/postal/message_parser.rb:

On line 4:

URL_REGEX = /(?<url>(?<protocol>https?)\:\/\/(?<domain>[A-Za-z0-9\-\.\:]+)(?<path>\/[A-Za-z0-9\.\/\+\?\&\-\_\%\=\,\~\:\;\(\)\[\]#]*)?+)/

I changed to:

URL_REGEX = /(?<url>(?<protocol>https?)\:\/\/(?<domain>[A-Za-z0-9\-\.\:]+)(?<path>\/[a-zA-Zà-úÀ-Ú0-9\.\/\+\?\&\-\_\%\=\,\~\:\;\(\)\[\]#]*)?+)/

This change works fine on regex101.com, but does not work on Postal. Actually it breaks all tracking urls.

[willpower232]

It looks like the implementation of regex in ruby might be a little different/opinionated

The documentation mentions some POSIX-style matching syntax which might be more accurate, are you able to check?

https://ruby-doc.org/core-2.7.0/Regexp.html#class-Regexp-label-Character+Properties